Reducing uncertainty in assessment of occupational diseases informs prevention

Estimating the contribution of workplace exposure to an individual worker’s overall risk of disease is a crucial step in the expert review of occupational diseases and involves a complex triad assessment of (i) clinical diagnosis, (ii) work-related risk factors and (iii) personal risk factors. The assessment of each of these three domains has its own uncertainties and should be evaluated in coherence, preferably in a multidisciplinary setting of a medical specialist, occupational physician and work exposure professional. While (quality of) evidence on disease aetiology from clinical, mechanistic and epidemiological studies has advanced, evaluating the specific attribution of work exposure to adverse health effects remains complex and is subject to scientific and societal debate in various countries.

In occupational health practice, we are witnessing a shift from well-recognized traditional occupational diseases, such as mesothelioma and silicosis, to an increased burden of multifactorial diseases, not only related to chemical agents, such as several cancers, lung and skin diseases, but also mental, musculoskeletal, infectious and neurological diseases. A major gap in the timely identification and reporting of occupational diseases is the lack of efficient, uniform registration systems with harmonized exposure and diagnostic criteria across countries. This hampers the collection of reliable exposure-disease data and eventually the development of targeted preventive measures. This editorial makes a plea for sustained scientific and professional focus to assess and report occupational diseases. A multi-evidence approach, integrating different streams of evidence, is essential to reduce uncertainty and improve the assessment of new and emerging occupational diseases.

Over the past decade, evidence synthesis in the field of causality has progressed significantly through data merging from large cohorts, systematic reviews, development of meta-analysis approaches, and the evaluation of study quality, resulting in an increase in confidence in identified relationships on the population level. For substance-related occupational (mostly cancers), increasing attention is given to harmonizing assessment criteria across different streams of evidence.¹ For mental and musculoskeletal disorders, there is a growing emphasis on harmonizing criteria for work-related diagnostics in conjunction with measurement instruments.^2,3 However, at the individual level, the risk of developing an occupational disease is also influenced by personal factors such as genetic predisposition, sex, age, lifestyle, and the presence of pre-existing conditions.

The evaluation of causality in occupational disease varies in complexity, depending on the certainty of both the diagnosis and the exposure assessment. The evaluation of monocausal diseases is relatively straightforward; for example, in cases of mesothelioma from asbestos exposure or nasal adenocarcinoma from high-level wood dust exposure, because the probabilities that the disease is caused by exposure in individuals with disease and exposure are generally above 85–95%. Determining the probability of causation introduces greater complexity and uncertainty for multicausal diseases. For the majority of multicausal diseases, establishing the proportion of occupational origin relies on dose-response relationships, obtained from occupational cohort studies. For a limited number of diseases, such as silicosis and occupational allergic asthma,^4,5 identifying the causative factor is an integral part of the clinical diagnosis; herewith reducing ­uncertainty in establishing the occupational origin, although differential diagnostic challenges remain. In their absence, a multidisciplinary expert judgement integrating scientific evidence, exposure information, biological and personal information, and clinical assessment becomes the last option. The varying complexity of the assessment, but also ways to reduce uncertainties in these assessments, will be illustrated with the following examples.

For occupational lung cancer caused by asbestos and silica, and occupational hearing loss, occupational knee arthrosis, dose-response relationships are available from cohort studies. A quantitative assessment of the probability that occupational exposure exceeds a cutoff value derived from these relationships often provides a basis for determining individual causality. ­Traditionally, a cutoff value is chosen that represents an exposure level which doubles the risk of disease (RR ≥ 2). However, these thresholds may differ between various countries due to differences in policy. Importantly, these relationships, based on population data, do not account for differences in individual susceptibility. Personal risk factors can act as confounders (e.g. smoking and lung cancer, high BMI and knee osteo arthritis) or as moderators of susceptibility (e.g. atopy in case of occupational allergy or pregnancy, increasing the risk of gestational diabetes and hypertensive disorders when occupational noise levels exceed LAeq8h 80 dB.^6,7). So, occupational physicians may consider using this information for selective prevention for vulnerable groups. However, how to ­interpret and incorporate these susceptibility factors into the assessment of individual causality with regard to registries and compensation remains a challenge and is still an open question.

Another example of reducing uncertainties in occupational origin is contact dermatitis (CD), a common occupational disease with two etiologically different forms: irritant and allergic, ICD and ACD, respectively.⁸ In ACD, identifying the causative factor is part of the clinical diagnosis (epidermal patch testing for identification of culprit allergen), and even a brief exposure to a low allergen dose could be sufficient to elicit disease. In contrast, ICD develops gradually, and an indication of a dose–response relationship for wet work, the most common risk factor for ICD, has been demonstrated.⁹ For ICD, in addition to information on exposure, the repetitive character of disease and disease course after exposure cessation is essential in determining causality. Although susceptibility plays a significant role, as a history of atopic dermatitis increases the likelihood of occupational ICD, its quantitative assessment to predict the probability of causation remains inconclusive. Thus, the primary approach to reducing uncertainty in assessing occupational origin differs between the two conditions: in ACD, clinical diagnosis and confirmation of exposure to the culprit allergen in the workplace are key, while in ICD, next to diagnosis and clinical course, exposure and history of atopic dermatitis are highly relevant. Next to establishing the occupational causality, selective and universal prevention can be informed.

For most mental disorders, there is increasing debate about the scientific basis for their recognition as occupational diseases. In the case of stress-related disorders, burnout or depression, neither the clinical diagnosis nor dose–response relationship can strengthen the certainty of classification as an occupational disease. There is evidence that high work pressure, an imbalance between high effort and low reward, low job autonomy, lack of colleague support, organizational injustice and workplace bullying contribute to the development of stress-related disorders or depression. Risk estimates, however, rarely indicate a single exposure level that doubles the risk of developing a mental disorder—a usual cut-point for the probability of causation, although a combination of risk estimates might.¹⁰ Moreover, personal factors as perception of health and vulnerability, attitudes towards work, educational attainment and personality may modify their risk of future mental illness.¹¹ Although the occupational origin of mental disorders is in debate in many countries, such as the recent planned ban of (occupational) burnout as a medical diagnosis in Sweden, reducing clinical and etiological uncertainties remains key to inform preventive action at organizational and individual levels.

Worldwide, countries have different approaches to establish the assessment of occupational diseases and therefore hindering the progress in reliable epidemiological data sources. Illustrative is the case of occupational skin cancer in outdoor workers caused by solar UVR, where the assessment of individual causality varies across countries. Some countries rely on a straightforward evaluation based on the type and duration of outdoor work, while others use more complex algorithms that account for both occupational and leisure UV exposure, with age as a proxy to estimate leisure exposure.¹² Susceptibility factors like skin phototype, however, are still not considered in the assessment of occupational skin cancers.

Ultimately, determining whether an occupational disease is present in an individual requires a well-reasoned and explicit assessment based on clinical diagnosis, along with sufficient understanding of biologically relevant work-related exposure and personal factors. Several personal factors have been shown to influence the risk, severity and prognosis of occupational diseases. Scientific knowledge of occupational risk factors—taking into account mediators (e.g. socioeconomic position), moderators (e.g. individual susceptibility) and confounders (e.g. exposure to nonoccupational factors)—can improve understanding of modifiable factors for the universal and selective prevention of occupational diseases.

Greater recognition of work as a risk-modifying factor in national and international occupational health and safety policies, medical guidelines and absenteeism regulations aligns with broader societal trends toward prevention and sustainability. However, the evidence base is limited, lacks harmonization and does not incorporate novel approaches to reduce uncertainties in the assessment of occupational diseases, thereby missing opportunities to contribute to the prevention of occupational and work-related diseases. This editorial aims to open the scientific and professional debate to shift towards the challenge of new and emerging multicausal diseases, where the assessment of occupational origin is made explicit by reducing uncertainties in coherence of clinical diagnosis and work-related and personal risk factors, but also stimulate multidisciplinary efforts in advanced longitudinal exposure monitoring (e.g. wearables and biomarkers) and diagnostic approaches for differential clinical diagnosis (e.g. use of biomarkers, detailed occupational histories).

COMPETING INTERESTS

None declared.